1653. Multi-enabler transportation network design for autonomous trucks with elastic demand
Invited abstract in session MD-57: Smart mobility: Exploring the Potential of Connected and Autonomous Vehicles, stream Transportation.
Monday, 14:30-16:00Room: Liberty 1.12
Authors (first author is the speaker)
| 1. | Ebrahim Mohammadi
|
| Rotterdam School of Management, Erasmus University Rotterdam | |
| 2. | Marie Schmidt
|
| Institute for Computer Science, University of Würzburg | |
| 3. | Rob Zuidwijk
|
| Technology Operations Management, RSM Erasmus University |
Abstract
Autonomous truck (AT) transportation has the potential to reduce transportation costs and improve road safety. The successful adoption of this technology relies on key enablers such as refueling stations, emergency maintenance facilities, and transfer hubs. However, the actual demand for ATs may be lower than the nominal demand, as customers may not immediately switch to autonomous transportation. The decision to adopt AT may depend on the cost-benefit it provides to customers, which, in turn, is influenced by network design decisions. In this paper, we investigate the problem of multi-enabler transportation network design for autonomous transportation with elastic demand. We assume that demand for ATs depends on the cost savings it offers to customers, and these savings are directly impacted by network design choices. We develop a mixed-integer mathematical model for locating infrastructure enablers required for long-haul autonomous transportation. To solve the model for larger instances, we developed a Column Generation based heuristic. The model is implemented numerically for benchmark cases, such as the Sioux Falls network, and a sensitivity analysis is conducted for key parameters. Additionally, the findings provide several managerial insights, such as the investment required to achieve specific adoption level.
Keywords
- Network Design
- Transportation
- Column Generation
Status: accepted
Back to the list of papers